Introduction

The elasmobranch fishes (sharks, skates, and rays) are primitive vertebrates that evolved at least 400 million years ago, and are considered to be the oldest living jawed vertebrates (Maisey 1980; Wilga et al. 2001). Many are considered important consumers and are hypothesized to play a major role in structuring marine ecosystems through predation (Carlson et al. 2004; Bethea et al. 2007). With this in mind there has been a great deal of interest in understanding their diet, rate of ingestion, and digestion within the past few decades. Therefore, many studies have focused on topics such as diet habits, tooth morphology and growth, stomach content, and digestive strategies (Wetherbee and Gruber 1993; Cortes et al. 1996; Lowe 2001; Bush and Holland 2002).

Despite extensive research on elasmobranch food habits and feeding behavior, there are virtually no published studies that have explored other important facets of digestion in these fish; for example, how dietary macromolecules are digested and absorbed across the gastrointestinal tract. In fact, detailed studies on some of these topics have only recently been conducted. For example, Jhaveri et al. (2015) recently examined digestive enzyme activity throughout the gastrointestinal tract of the bonnethead (Sphyrna tiburo), demonstrating extensive enzymatic activity in the scroll valve intestine, confirming longstanding but largely uncorroborated beliefs that it is the most nutrient absorptive component of the shark gut. To date, the only work that has focused on the actual mechanisms of nutrient uptake in the elasmobranch digestive tract was a study conducted by Crane et al. (1979), which demonstrated the presence of potential d-glucose transporters in the brush border membrane (BBM) of small dogfish (Scyliorhinus canicula) spiral valve intestine. That study showed that glucose uptake by dogfish spiral valve BBM is stimulated by a Na+ gradient, suggesting that, as in other vertebrate intestines, secondary active transport is the primary mechanism of glucose uptake. However, no studies have investigated the uptake of products of protein digestion in elasmobranchs despite their largely carnivorous feeding habits. This is unfortunate because as the oldest living vertebrates to exhibit digestive tract modifications associated with jaw evolution, knowledge about peptide uptake in elasmobranchs will expand our understanding of the evolution of this process in vertebrates.

It is known that dietary proteins can be degraded into free amino acids within the vertebrate digestive tract. However, a great deal of studies provide evidence that within the vertebrate intestine, many dietary proteins are broken down into di- and tripeptides rather than single amino acids (Clements and Raubenheimer 2006; Verri et al. 2010). The uptake of di- and tripeptides is mediated by integral plasma membrane transport proteins that belong to the peptide transporter family (PTR family). Described vertebrate peptide transporters include peptide transporters 1 (PepT1) and 2 (PepT2). Specifically, PepT1 is a Na+–independent, H+ –dependent cotransporter protein present in the intestinal BBM that is responsible for the translocation and absorption of di- and tripeptides released during digestion by luminal and membrane-bound proteases. PepT2, however, is predominately located in the kidney BBM, brain, gonads, lungs, and gallbladder (Leibach and Ganapathy 1996; Saito et al. 1996). To date, no studies have investigated the presence of these important transport proteins in the elasmobranch gastrointestinal (GI) tract.

As there are consistencies in the distribution of PepT1 and PepT2 across vertebrates, we expected to find elasmobranch PepT1 to have the same range of expression specifically in the spiral valve intestine and other organs of the GI tract. Therefore, the purpose of this study was to isolate and sequence pept1 from the elasmobranch spiral valve intestine, and to histologically assess the distribution of this transporter in all components of the GI tract and the pancreas. In addition, this study examined the uptake capabilities of this intestinal peptide transporter. For this particular study bonnethead sharks were used as the model species due to their abundance and availability.

Materials and methods

Sample collection and tissue preparation

Bonnethead sharks were collected from sites ranging from Charleston, South Carolina to Cape Canaveral, Florida using gillnet and longline fishing. Animals were euthanatized using IACUC-approved methodology, and the gastrointestinal tract was obtained by dissection (Fig. 1). Small portions of the scroll valve intestine were stored in RNAlater at −4 °C for use in molecular studies. Samples (~2–3 mm) of the gastrointestinal tract and accessory organs (esophagus, stomach, pancreas, duodenum, scroll valve intestine, and rectum) were also obtained and fixed in 10 % formalin in elasmobranch-modified saline for approximately 48 h, then rinsed and transferred to 70 % ethanol for long-term storage until used for histology and immunohistochemistry. Last, whole intestines were obtained from some sharks and stored at −80 °C until used for vesicle experiments.

Fig. 1
figure 1

Bonnethead and bonnethead gastrointestinal tract, demonstrating morphology of all components used in the present study

Full size image

Isolation of S. tiburo pept1 and pept2 cDNA sequences

To isolate total RNA, approximately 50 mg of scroll valve intestine was minced, placed in 750 µL of TRI Reagent (Zymo Research, Irvine, CA) and homogenized. The homogenate was centrifuged for 1 min at 12,000 g to pellet any remaining solid tissue, with 700 µL of the supernatant subsequently placed into a microcentrifuge tube. The Direct-Zol RNA Miniprep kit (Zymo Research, Irvine, CA) was then used for total RNA extraction. The concentration and purity of RNA was determined using a NanoDrop spectrophotometer and gel electrophoresis. Reverse transcription of 1 µg of total RNA into cDNA was conducted using Superscript III reverse transcriptase (Life Technologies) following the manufacturer’s instructions.

For degenerate PCR, degenerate primers were designed based on conserved portions of pept1 sequences from chicken (NM_204365.1), salmon (NM_001146682.1), eel (AB762417.1), and zebrafish (NM_198064.1). Four primer combinations were created by designing two forward and two reverse primers, to maximize the possibility of at least one combination annealing to the cDNA and amplifying a portion of pept1. PCR was performed using six different cycling parameters, to increase the range of annealing temperatures tested. Cycling parameters included: 95 °C for 2 min followed by initial annealing temperatures of 52.5, 53.3, 54.9, 57.2, 60.1 or 62.5 °C (a different starting temperature for each reaction) for 30 s and then 72 °C for 1 min. For the first 25 cycles, the annealing temperature was decreased each cycle by 0.5 °C, followed by 15 cycles (for 40 total) at an annealing temperature of 40 °C for all reactions. Following PCR, products were examined by electrophoresis on a 1 % agarose gel to determine the optimal temperature and combination of primers, with products of the appropriate size isolated, cloned and sequenced as described below.

The primers that yielded the largest products were 5′-GAGTTCTGYGARMGDTTCTCCTACT-3′ as the forward primer and 5′-TGGTCAAANARDGYCCAGAACAT-3′ as the reverse primer. Products from PCR reactions using this primer combination were ligated into the pGEM-T Vector System I and used to transform competent Escherichia coli. Twenty-five µL of transformed cells were then plated on agar plates coated with isopropyl-beta-D-thiogalactopyranoside (IPTG) and 5-bromo-4-chloro-3-indolyl-beta-D-galacto-pyranoside (X-gal) for blue–white screening and incubated at 37 °C overnight. White (positive) colonies were then selected, placed in 2 mL of LB broth and incubated overnight at 37 °C with shaking at 250 rpm. Overnight incubations were transferred into capped tubes and centrifuged at high speed for 30 s. The supernatant was decanted and pellet resuspended in 600 µL of sterile water. The Zippy Plasmid Miniprep Kit (Zymo Research) was then used to isolate plasmid DNA from the full 600 µL for sequencing. Selected plasmids were sequenced by GENEWIZ (South Plainfield, NJ), and sequences were analyzed using the CLC Main Workbench 7 software (CLC bio, Qiagen). Although the target cDNA was pept1, the methods described above yielded clones positive for both pept1 and pept2 sequences, likely due to high sequence similarity within conserved regions used for degenerate primer design; therefore full-length cDNA sequences were obtained for both genes using rapid amplification of cDNA ends (RACE) as described below.

To obtain the remaining 3′ and 5′ coding sequence as well as the 5′ and 3′ untranslated regions of pept1 and pept2, the FirstChoice RLM-RACE kit (Life Technologies) was used. Total RNA from scroll valve intestine was extracted as described above, and 5′ and 3′ RACE-ready cDNA was generated following the manufacturer’s instructions. 5′ and 3′ RACE PCR products were obtained using nested PCR reactions that included kit-specific primers and species-specific primers (Table 1) for pept1 and pept2, designed using the cDNA fragments isolated as described above. RACE products were cloned and sequenced as described above, and sequences were assembled using CLC Main Workbench 7.

Table 1 Species-specific primers used to obtain the remaining 5′ and 3′ coding sequence and untranslated regions from bonnethead PEPT1 and PEPT2
Full size table

Phylogenetic analyses

BLAST analysis was used to confirm the identity of the presumed S. tiburo pept1 and pept2 sequences (http://www.ncbi.nlm.nih.gov/BLAST). S. tiburo sequences were then aligned with pept1 and pept2 sequences from other taxa using the Clustal W algorithm in CLC Main Workbench for analysis of gene homology. Potential transmembrane domains for both PepT1 and PepT2 were defined using TMprep computational resources (http://www.bioinformatics.utep.edu/BIMER/tools/transmembrane.html). For phylogenetic analysis, both S. tiburo protein sequences were aligned with PepT1 and PepT2 sequences from diverse vertebrate taxa using the Clustal W algorithm in MEGA version 6 (Tamura et al. 2013); phylogenetic relationships were determined using the Neighbor-Joining method with 2000 iterations to generate a bootstrap consensus tree.

Immunohistochemistry

Fixed organ samples were processed for routine paraffin histology as described by Gelsleichter et al. (2003). Once embedded in paraffin, samples were sectioned (5 µm) using a rotary microtome and mounted on poly-l-lysine-coated slides. Immunohistochemistry was performed to examine the cellular location of PepT1 using rabbit polyclonal anti-rat PepT1 (SLC15a1, Millipore, Berlin, Germany) and the Vector ImmPRESS anti-rabbit kit (Vector Laboratories, Burlingame, CA). Tissue sections were incubated in a limonene-based solvent (Citrisolv) for deparaffinization, rehydrated in a descending series of graded alcohol concentrations (100–95 %), and then rinsed for 10 min in a running tap water bath. Rinsed sections were incubated in an antigen retrieval solution (10 mM sodium citrate, pH 6.0) at 95 °C for 20 min to expose any epitopes of the target antigen that may have been masked by the fixation process. After cooling to RT, sections were then rinsed in reverse osmosis (RO) water, followed by phosphate buffered saline (PBS), and then blocked for nonspecific reactivity with primary antibodies by overnight incubation in 2 % normal goat serum in PBS (Vector Laboratories) at 4 °C. After blocking, slides were washed with PBS and endogenous peroxidase activity was quenched by incubation in a 1:1 mixture of 3 % hydrogen peroxide and methanol for 15 min. Sections were then rinsed again in two separate baths of PBS. Sections were incubated at 4 °C overnight in primary antibody diluted 1/1000 in a PBS solution containing 0.1 % gelatin and 0.1 % sodium azide (G-PBS). Negative control sections were incubated in diluent only. Following incubation, sections were rinsed in a PBS bath containing 0.05 % TWEEN-20 (PBS-T), followed by two additional PBS rinses. Afterwards, slides were incubated for approximately 30 min with secondary antibody, anti-rabbit Ig-HRP conjugate. Following this incubation, sections were rinsed in three separate PBS baths and then incubated in the chromogen 3.3′-diaminobenzidine (DAB), using the ImmPACT DAB Peroxidase (HRP) substrate kit (Vector Laboratories). Slides were rinsed in tap water and then counterstained in 2 % methyl green (Vector Laboratories) for 15–60 min at 37 °C. Afterwards, they were rinsed briefly in tap water, dehydrated in an ascending series of graded alcohols (95–100 %), cleared using Citrisolv, and mounted using Cytoseal 60 (Thermo Scientific, Fair Lawn, NJ). Microscopy was used to examine the distribution of immunoreactive PepT1 in sections.

Preparation of brush border membrane vesicles

Whole intestines stored at −80 °C were thawed on ice in PBS. Subsamples of intestine were cut and scraped using a razor blade to free epithelial cells into 60 mL of a 300 mM mannitol, 20 mM Tris HCl, 50 mM EGTA, 1 mM PMSF buffer adjusted to pH 7.0 (buffer 1). For BBM vesicle (BBMV) purification, a method used on Mozambique tilapia (Oreochromis mossambicus) intestine (Reshkin and Ahearn 1987; Thamotharan et al. 1996) was used. This method yields BBMV that are enriched more than tenfold in brush border enzymes compared to the original tissue homogenate. Similar methodology was used by Crane et al. (1979) for studies on S. canicula, demonstrating that elasmobranch BBMV can be isolated using this approach. The expanded technique has two additional washing steps for discarding stored cytoplasmic digestive enzymes (Ahearn and Storelli 1994). Purified BBMV were then used for transport measurements.

Transport measurements

Transport experiments were conducted using the scroll intestine BBMV and the Millipore filtration technique (Thamotharan et al. 1996). [14C] Glycylsarcosine (Gly-Sar) was used in 120 min uptake experiments by mixing 20 µL of membrane suspension with 180 µL of radiolabelled incubation medium. Composition of incubation medium varied with the nature of each experiment. Effect of pH on uptake of radiolabeled Gly-Sar in intestinal BBMV was examined using incubation media containing 150 mM KCl and 20 mM HEPES adjusted to a pH of 7.5 or a pH of 8.5, and a third medium containing 150 mM KCl and 20 mM MES adjusted to pH 5.5. All vesicles were loaded with 150 mM KCl and 20 mM HEPES adjusted to pH 7.5. Measurements were taken at intervals of 0.25, 1, 2, 5, 10, 60, and 120 min.

Kinetics of [14C] Gly-Sar influx at 1 min was then examined using media composed of 150 mM KCl, 20 mM HEPES at concentrations of 1, 2.5, 5, 10, and 25 mM Gly-Sar.

In these two separate experiments, uptake of [14C] Gly-Sar was terminated by injecting 20 µL of the reaction into 2 mL of a stop solution (composed of same constituents as the incubation medium without the radiolabeled dipeptide). Solution was then filtered using a Millipore filter (0.65 µm pore diameter) and washed with an additional 3 mL of stop solution. Filters were then placed in 3 mL of Beckman Liquid scintillation cocktail and counted using a Beckman LS-6500 scintillation spectrometer. The averages of each set of replicates were determined and graphed using SigmaPlot 10.0. One-way ANOVA was used to determine if differences between the mean values of Gly-Sar uptake among the given ion gradients where the overshoot was present were statistically significant.

Results

The bonnethead pept1 cDNA was 4055 bp, with an Open Reading Frame (ORF) of 2,157 bp, encoding a putative protein of 718 aa. The sequence also included 455 bp of 5′ untranslated region (UTR) sequence and 1443 bp of 3′UTR sequence with a polyadenosine mRNA tail. Hydropathy analysis predicted 12 potential transmembrane domains (TMD) with an extracellular loop between TMD IX and X (Fig. 2). When compared to pept1 sequences from other vertebrates using BLAST, the bonnethead sequence showed high sequence identity, ranging from 60 to 67 %.

Fig. 2
figure 2

Amino acid sequence of bonnethead PepT1 aligned with PepT1 proteins from diverse vertebrate taxa: Danio rerio, Gallus gallus, and Rattus norvegicus. Lined region with roman numerals indicative of putative TMDs

Full size image

Pept2 cDNA was also isolated from the bonnethead shark intestine. The sequence was 2549 bp, with an ORF of 2169 bp, encoding a putative protein of 722 aa. The sequence included 91 bp of 5′ UTR sequence and 289 bp of 3′UTR sequence with a polyadenosine mRNA tail. Hydropathy analysis predicted 13 potential TMD with an extracellular loop between TMD X and XI (Fig. 3). When compared to pept2 sequences from other species using BLAST, the bonnethead sequence showed high sequence identity, ranging from 62 to 66 %.

Fig. 3
figure 3

Amino acid sequence of bonnethead PepT2 aligned with PepT2 proteins from diverse vertebrate taxa: Danio rerio, Gallus gallus, and Rattus norvegicus. Lined region with roman numerals indicative of putative TMDs

Full size image

Percent identity between the two isolated intestinal peptide transporters was only 51 %, compared to much higher identity between PepT1 and PepT2 proteins within mammalian species (~80 %; Wang et al. 2010). Also, the phylogenetic reconstruction of the vertebrate peptide transporter proteins assigned the two isolated bonnethead shark peptide sequences to separate clades. Within the PepT1 and PepT2 branches of the phylogenetic tree, early divergence of the elasmobranch protein sequences from those of the teleost and mammalian groups was predicted (Fig. 4).

Fig. 4
figure 4

Phylogenetic analysis of pept1 (slc15a1) and pept2 (slc15a2). Relationships were inferred using the Neighbor-Joining method in Mega6 (Tamura et al. 2013). pept1 analysis includes sequences from Callorhinchus milii (XP_007904487.1), Meleagris gallopavo (NP_001290095.1), Gallus gallus (NP_989696.1), Rattus norvegicus (NP_476462.1), Mus musculus (NP_444309.2), Bos taurus (NP_001092848.1), Homo sapiens (NP_005064.1) Latimeria chalumnae (XP_005992366.1) Anguilla japonica (BAM67012.1), Xenopus tropicalis (XP_002935692.2), Cyprinus carpio (AEX13747.1), Epinephelus aeneus (AFP33141.1), Larimichthys crocea (NP_001290295.1), and Danio rerio (NP_932330.1), Salmo salar (NP_001140154.1). pept2 analysis includes sequences from Mus musculus (NP_067276.2), Homo sapiens (NP_066568.3), Gallus gallus (AGZ02797.1), Danio rerio (NP_001034917.1), Xenopus laevis (NP_001080398.1), Larimichthys crocea (KKF11892.1), Oryzias latipes (XP_004081581.1), Cyprinus carpio (ADM48102.1), Latimeria chalumnae (XP_006004055.1), Sus scrofa (NP_001090983.1), Rattus norvegicus (NP_113860.2), Meleagris gallopavo (XP_010712188.1), and Callorhinchus milii (XP_007907469.1). Outgroup from Ciona intestinalis (XP_002121251.1:pept2-slc15a2). Numbers at branch points indicate the percentage of 2000 bootstrap replicates supporting the division

Full size image

Immunohistochemistry was conducted on all components of the bonnethead gastrointestinal system from 10 individuals and PepT1 was detected in multiple areas. Staining of such organs revealed the epithelium of the esophagus, stomach, duodenum, scroll valve intestine, rectum and pancreatic acinar cells as distinctly immunopositive, implying the presence of PepT1. This staining was relatively diffuse throughout the cytoplasm, suggesting that active cellular PepT1 synthetic and transfer processes appeared to be taking place throughout the digestive organs. However, staining of the stomach and scroll valve intestine appeared more intense than that of the other gastrointestinal organs (Fig. 5b). Moderate positive staining was detected within the duodenum, rectum, and pancreatic acinar cells in comparison to staining within the stomach and scroll valve intestine (Fig. 5b). There was also minimal immunopositive staining observed within the esophagus, as only minor distinct differences were observed between the control and non-control slides (Fig. 5a, b). Because results based only on immunohistochemistry are difficult to quantify, other more semi-quantitative methods such as Western blot or functional assays of transport activity from the different regions might provide additional localization support for the present findings.

Fig. 5
figure 5

Results from immunohistochemical analysis of PepT1 in transverse sections of multiple components of the bonnethead gastrointestinal tract. Location of sections is provided on the left margin. Results of negative controls are displayed in column a. Arrows represent positive immunostaining

Full size image

Figure 6 illustrates the effects of a pH gradient on the time course of 1 mM [14C] Gly-Sar uptake by the bonnethead scroll valve BBMV. This time course was characterized by an overshoot at 1 min that was highest when the pH inside was 7.5 and the pH outside was 5.5 (Fig. 6). No overshoots were observed when the outside pH was 7.5 or 8.5.

Fig. 6
figure 6

Results from time course experiments, testing the effect of pH on 1 mM 3H-glycylsarcosine uptake in BBMV from bonnethead shark intestine. Vesicles were preloaded with 150 mM KCl, 20 mM HEPES at pH 7.5, and incubated in various pH levels (5.5, 7.5, 8.5). Experiments were conducted three times with three replicates each. Symbols are mean ± SE

Full size image

Influxes (1 min uptakes) of [14C] Gly-Sar into the scroll valve BBMVs were measured over a concentration range of 1–25 mM [14C] Gly-Sar in the presence of an inwardly-directed proton gradient (inside pH 7.5 and outside pH 5.5). Influxes for this concentration range best fit the Michaelis–Menten kinetics equation (J oi = {(J max × [S])/(K m + [S])}), where J oi is peptide influx, J max is maximal influx, K m is the Gly-Sar concentration at ½ J max, and S is Gly-Sar concentration. This analysis yielded Michaelis–Menten values for a saturable, low-affinity system with a K m = 6.18 ± 1.66 mM and a J max = 6077.86 ± 632.57 pmol/mg protein x min (n = 3) (Fig. 7). Curve fits using alternative equations involving two separate carriers or one carrier plus diffusion did not provide as good a fit to the data as assuming a single saturable carrier system. These results suggest that the influx contribution of a second carrier or diffusion to the observed influx data were minor compared to transport by the predominant process.

Fig. 7
figure 7

Results from kinetics experiment, testing the effect of concentration on 3H-glycylsarcosine uptake in BBMV from bonnethead shark intestine. Vesicles were preloaded with 150 mM KCl, 20 mM HEPES at pH 7.5, and incubated in 150 mM KCl, 20 mM MES at pH 5.5 solutions with various concentrations of 3H-glycylsarcosine (1, 2.5, 5, 10, 25 mM). Experiments were conducted three times with five replicates each. Symbols are mean ± SE

Full size image

Discussion

Multiple aspects of this study, including molecular characterization, immunohistochemistry, and BBMV experiments support the identification of cDNAs encoding two individual S. tiburo peptide transporters of the PTR family. The two encoded proteins, designated as PepT1 and PepT2, represent the products of orthologous genes to pept1 and pept2 of other vertebrate taxa. When compared to other known members of the PTR family, the predicted bonnethead PepT1 and PepT2 share significantly higher overall identity to other known PepT1 and PepT2 sequences, respectively, and are assigned to the alike monophyletic groups of the reconstructed phylogenetic tree. The analysis of domains also reveals that the bonnethead PepT1, like other vertebrate PepT1 proteins, includes 12 TMDs and the overall major area of difference lies within the large extracellular loop between TMDs 9 and 10. However, unlike PepT2 in other vertebrates, the bonnethead PepT2 sequence includes 12 strong and 1 weak predicted TMDs rather than just 12 strong TMDs (Wang et al. 2010) and the overall major area of difference lies within the large extracellular loop between TMDs 10 and 11 instead of between TMDs 9 and 10. However, because TMD 1 is very weak with a score of 524 it is possible that this TMD is not a major structural component and that there are in fact 12 TMDs with an extracellular loop between TMDs 9 and 10 as seen in other vertebrate PepT2 proteins. This scenario is further supported by the presence of a lysine residue in what would then be TMD 7, as discussed below.

PepT1 in the bonnethead shark was localized primarily in the epithelial cells of multiple gastrointestinal organs. The presence of this transporter within the stomach, duodenum, scroll valve intestine, and rectum gives insight into the absorptive qualities of such organs. With PepT1 present in the epithelial lining of these organs, it can be concluded that in each of these structures, there is some level of absorption of dietary peptides as they digest their prey throughout the entire gastrointestinal tract. Also, the minimal expression of PepT1 within the esophagus provides further support that elasmobranchs typically engulf their prey whole and therefore there is minimal need for peptide absorption in the esophagus, with initial breakdown instead occurring within the stomach. The elasmobranch stomach is the first organ used for major digestion through the release of hydrochloric acid (HCl), which converts the inactive zymogen pepsinogen into the active protease enzyme pepsin, initiating the digestion and absorption of proteins (Papastamatiou and Lowe 2004, 2005, Papastamatiou 2007). Within the stomach, prey is processed into chyme, an acidic fluid consisting of gastric juices and partly digested food, and then passed into the intestine (Camilleri et al. 1986). Once chyme enters the shark spiral/scroll valve intestine, it has been determined that pancreatic enzyme activities largely decrease moving down the gut while BBM enzyme activities peak, suggesting that the spiral/scroll valve intestine is the primary site of absorption (Jhaveri et al. 2015). Therefore, the strong immunoreactivity of PepT1 within the stomach and scroll valve intestine epithelial lining is likely correlated with important roles in dietary peptide absorption within these organs. Unfortunately, the localization of PepT2 was not explored in this study as it was not expected that the pept2 sequence would be isolated. However, the differential distribution of these two peptide transporters has been explored in a number of vertebrates, and PepT1 is characterized as mainly the intestinal peptide transport system (Verri et al. 2003, 2010) whereas PepT2 is characterized as mainly a renal peptide transporter (Daniel et al. 1991; Leibach and Ganapathy 1996; Saito et al. 1996). However, both transporters may co-occur in the same tissues or organs. For instance, both PepT1 and PepT2 are found within the mammalian renal proximal tubules (Leibach and Ganapathy 1996). It is thought that this organ may play a significant role in conserving peptide-bound amino acids and amino nitrogen that may otherwise be lost in urine via the peptide transport process. Therefore, the presence of both PepT1 and PepT2 may maximize the amount of peptides conserved before food exits the body (Matthews 1975; Schlagheck and Webb 1984; Seal and Parker 1991; Gardner 1994). With this in mind, unlike the teleost and mammalian intestine, the presence of both pept1 and pept2 mRNA within the bonnethead scroll valve intestine may indicate that both proteins are critical for absorbing the necessary amount of peptides needed for survival. The scroll valve intestine has been described as the most active and absorptive section of the shark intestinal system (Jhaveri et al. 2015) due to its unique structure, which conserves space within the body cavity by the infolding of the mucosa and submucosa in a spiral or scroll-like fashion. It may be necessary for elasmobranchs, which are known to consume large meals, to hold those meals for an extended period of time in the stomach (Wetherbee et al. 1987; Holmgren and Nilsson 1999; Papastamatiou 2007) and to have multiple active peptide transporters within the intestine to maximize nutrient absorbtion. An alternative suggestion for the co-occurance of PepT1 and PepT2 in bonnethead digestive organs is the possible expression of PepT1 in absorptive epithelial cells and PepT2 localization in enteric nervous system cells, the latter following the described pattern in mammalian systems (Rühl et al. 2005).

It is unknown at the present time what the function of PepT1 in the pancreatic acinar cells might be. However, it is interesting to note that PepT1 has also been detected in cells of the exocrine pancreas in other vertebrates such as humans in past studies (Bockman et al. 1997). Interestingly, human pancreatic cancer cells have been shown to overexpress PepT1 (Gonzales et al. 1998) and both agonists and antagonists of this transporter have been shown to be capable of inhibiting the growth of these cells (Mitsuoka et al. 2010), suggesting that PepT1 may be a useful target for inhibiting the progression of pancreatic cancer.

The functional aspects of mammalian peptide transporters have been well-examined in the past decade using BBMV techniques, which have allowed detailed characterization of the kinetics and ion- dependent properies of such transporters. The functional results from this study provide new insight into the mechanism and driving force for dipeptide transport in a shark scroll valve intestine. These data show that the uptake of [14C] Gly-Sar in S. tiburo was stimulated by a proton gradient, in the absence of sodium. This is further supported by the presence of a lysine amino acid midway through TMD 7, specifically at amino acid 294 (PepT1) and at amino acid 297 (PepT2) in the bonnethead shark. Meredith (2009) found that the mutation of this residue to anything other than a positively charged residue (such as arginine or lysine) abolishes the stimulation of transport by a proton electrochemical gradient. This author identified that the loss of this positively charged residue is linked to the stoichiometry (1 proton: 1 dipeptide) of a proton coupled transport system. Therefore, rather than a true channel being formed, to allow peptide transit, without the lysine residue in the appropriate molecular location there is a small slippage of ion during the conformational change which occurs and prevents the translocation of the peptide to the other side of the membrane (Meredith 2009).

Gly-Sar uptake by intestinal BBMV in the bonnethead shark appears to be mediated by a low-affinity, high-capacity type carrier system. This low-affinity carrier system exhibited consistent quantitative K m kinetic constant binding values with those reported in mammalian and teleost studies, which ranged from 0.2 to 10 mM (Thamotharan et al. 1996; Skopicki et al. 1991).

In conclusion, the isolation of both pept1 and pept2 mRNA from the bonnethead scroll valve intestine provides a new understanding of the elasmobranch gastrointestinal system and insight into the absorptive capabilities of this unique organ. With this information along with the distribution and functional qualities of the PepT1 protein, we can surmise that the scroll valve intestine, stomach, duodenum, and rectum all appear to play significant roles in peptide absorption. Such information will lead to a better understanding of individual animal’s foraging behavior, growth and development, and evolutionary potential, and overall better grasp their ecological role.